Adaptive perception with locally-adaptable sensor array

Róbert Wagner, Ákos Zarándy, Tamás Roska

Research output: Contribution to journalConference article

Abstract

In this paper we propose some biologically motivated local image sensor adaptation methods. The special feature of these methods is that they change the integration time of the imager on the pixel level. Using these methods, low dynamic range integration type CMOS sensors will be able to perceive high dynamic range scenes, and at the same time, compress their dynamic range by keeping high contrast and without introducing non-existing edges.

Original languageEnglish
Pages (from-to)159-164
Number of pages6
JournalIEEE International Conference on Neural Networks - Conference Proceedings
Volume1
Publication statusPublished - Dec 1 2004
Event2004 IEEE International Joint Conference on Neural Networks - Proceedings - Budapest, Hungary
Duration: Jul 25 2004Jul 29 2004

    Fingerprint

Keywords

  • Bio-inspired visual sensor adaptation
  • High dynamic range image sensors
  • Locally adaptive image perception

ASJC Scopus subject areas

  • Software

Cite this